A major goal of evolutionary genetics is to understand how selection shapes patterns of DNA sequence variation, and ultimately, to connect DNA sequence variants to phenotypes that affect fitness. One promising approach to this problem is to study species that have recently expanded their ranges into new environments, to look for correlated changes in allele frequencies and environmental variables. This approach to studying environmental adaptation has been used successfully in Drosophila melanogaster where clinal variation is seen for a variety of different molecular and phenotypic traits across broad latitudinal transects. Recent work on humans has also linked changes in allele frequency to climatic variables in candidate genes for common metabolic disorders, suggesting that this approach may also allow us to identify important disease genes. The house mouse, Mus musculus, provides a unique opportunity to study environmental adaptation in the best mammalian model for humans. The proposed research will sample 260 wild mice representing 26 populations across a range of latitudes and altitudes in both North and South America. The sampling design includes paired transects for both altitude and latitude. All mice will be genotyped using a new Affymetrix SNP chip to identify candidate regions of the genome harboring genes underlying adaptation to different environments and climates. Select genomic regions will then be resequenced to identify allelic variants. Mice offer specific opportunities for functional tests not available in humans. Thus, the proposed work will also establish 16 new inbred strains of mice from extreme environments, sequence their genomes, and provide a preliminary description of their phenotypes. These resources will set the stage for future work linking particular adaptive variants to specific phenotypes, many of which are likely to be relevant for human metabolic disorders.